Classification-Based Anomaly Detection for General Data

About

Anomaly detection, finding patterns that substantially deviate from those seen previously, is one of the fundamental problems of artificial intelligence. Recently, classification-based methods were shown to achieve superior results on this task. In this work, we present a unifying view and propose an open-set method, GOAD, to relax current generalization assumptions. Furthermore, we extend the applicability of transformation-based methods to non-image data using random affine transformations. Our method is shown to obtain state-of-the-art accuracy and is applicable to broad data types. The strong performance of our method is extensively validated on multiple datasets from different domains.

Liron Bergman, Yedid Hoshen · 2020

Related benchmarks

Task	Dataset	Result
Anomaly Detection	SWaT	F1 Score—	348
Anomaly Detection	WBC	ROCAUC0.995	151
Anomaly Detection	CIFAR-10	AUC88.2	136
Out-of-Distribution Detection	ImageNet	AUROC91.8	113
Out-of-Distribution Detection	CIFAR-100	AUROC77.2	107
Anomaly Detection	MNIST	AUC92.01	87
Tabular Anomaly Detection	pima	AUC ROC0.718	86
Tabular Anomaly Detection	BreastW	AUC-ROC0.8328	83
Tabular Anomaly Detection	Wine	AUC-ROC0.8343	72
Tabular Anomaly Detection	pendigits	AUC-ROC86.5	72